Dueling Bandits with Qualitative Feedback


  • Liyuan Xu The University of Tokyo
  • Junya Honda The University of Tokyo
  • Masashi Sugiyama The University of Tokyo




We formulate and study a novel multi-armed bandit problem called the qualitative dueling bandit (QDB) problem, where an agent observes not numeric but qualitative feedback by pulling each arm. We employ the same regret as the dueling bandit (DB) problem where the duel is carried out by comparing the qualitative feedback. Although we can naively use classic DB algorithms for solving the QDB problem, this reduction significantly worsens the performance—actually, in the QDB problem, the probability that one arm wins the duel over another arm can be directly estimated without carrying out actual duels. In this paper1, we propose such direct algorithms for the QDB problem. Our theoretical analysis shows that the proposed algorithms significantly outperform DB algorithms by incorporating the qualitative feedback, and experimental results also demonstrate vast improvement over the existing DB algorithms.




How to Cite

Xu, L., Honda, J., & Sugiyama, M. (2019). Dueling Bandits with Qualitative Feedback. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 5549-5556. https://doi.org/10.1609/aaai.v33i01.33015549



AAAI Technical Track: Machine Learning